# Scaling Laws For Diffusion Transformers

arXiv: https://arxiv.org/abs/2410.08184 · HTML: https://mlsanity.com/paper/2410.08184/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 53 of its 63 references that are in the map.
- First posted on arXiv 10 Oct 2024.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 30 of 53)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Deep Unsupervised Learning using Nonequilibrium Thermodynamics](https://mlsanity.com/paper/1503.03585/) (2015)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [Generative Modeling by Estimating Gradients of the Data Distribution](https://mlsanity.com/paper/1907.05600/) (2019)
- [Neural Ordinary Differential Equations](https://mlsanity.com/paper/1806.07366/) (2018)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [Density estimation using Real NVP](https://mlsanity.com/paper/1605.08803/) (2016)
- [Glow: Generative Flow with Invertible 1x1 Convolutions](https://mlsanity.com/paper/1807.03039/) (2018)
- [Root Mean Square Layer Normalization](https://mlsanity.com/paper/1910.07467/) (2019)
- [Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Models](https://mlsanity.com/paper/1505.04870/) (2015)
- [Variational Diffusion Models](https://mlsanity.com/paper/2107.00630/) (2021)

## Built on by (most cited first, 30 of 44)

- [SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer](https://mlsanity.com/paper/2501.18427/) (2025)
- [Seaweed-7B: Cost-Effective Training of Video Generation Foundation Model](https://mlsanity.com/paper/2504.08685/) (2025)
- [xDiT: an Inference Engine for Diffusion Transformers (DiTs) with Massive Parallelism](https://mlsanity.com/paper/2411.01738/) (2024)
- [ScaMo: Exploring the Scaling Law in Autoregressive Motion Generation Model](https://mlsanity.com/paper/2412.14559/) (2024)
- [Scaling Laws for Optimal Data Mixtures](https://mlsanity.com/paper/2507.09404/) (2025)
- [MarDini: Masked Autoregressive Diffusion for Video Generation at Scale](https://mlsanity.com/paper/2410.20280/) (2024)
- [Reflect-DiT: Inference-Time Scaling for Text-to-Image Diffusion Transformers via In-Context Reflection](https://mlsanity.com/paper/2503.12271/) (2025)
- [TREAD: Token Routing for Efficient Architecture-agnostic Diffusion Training](https://mlsanity.com/paper/2501.04765/) (2025)
- [Towards Precise Scaling Laws for Video Diffusion Transformers](https://mlsanity.com/paper/2411.17470/) (2024)
- [Gemstones: A Model Suite for Multi-Faceted Scaling Laws](https://mlsanity.com/paper/2502.06857/) (2025)
- [Normalized Attention Guidance: Universal Negative Guidance for Diffusion Models](https://mlsanity.com/paper/2505.21179/) (2025)
- [LiT: Delving into a Simple Linear Diffusion Transformer for Image Generation](https://mlsanity.com/paper/2501.12976/) (2025)
- [How far can we go with ImageNet for Text-to-Image generation?](https://mlsanity.com/paper/2502.21318/) (2025)
- [DiT-Air: Revisiting the Efficiency of Diffusion Model Architecture Design in Text to Image Generation](https://mlsanity.com/paper/2503.10618/) (2025)
- [Turbo-GS: Accelerating 3D Gaussian Fitting for High-Quality Radiance Fields](https://mlsanity.com/paper/2412.13547/) (2024)
- [AERIS: Argonne Earth Systems Model for Reliable and Skillful Predictions](https://mlsanity.com/paper/2509.13523/) (2025)
- [Boogu-Image-0.1: Boosting Open Agentic Multimodal Generation via Understanding under a Minimal Budget](https://mlsanity.com/paper/2607.13125/) (2026)
- [How to Upscale Neural Networks with Scaling Law? A Survey and Practical Guidelines](https://mlsanity.com/paper/2502.12051/) (2025)
- [Diagnosing and Improving Diffusion Models by Estimating the Optimal Loss Value](https://mlsanity.com/paper/2506.13763/) (2025)
- [Scalable GANs with Transformers](https://mlsanity.com/paper/2509.24935/) (2025)
- [T-SHRED: Symbolic Regression for Regularization and Model Discovery with Transformer Shallow Recurrent Decoders](https://mlsanity.com/paper/2506.15881/) (2025)
- [Scaling Laws are Redundancy Laws](https://mlsanity.com/paper/2509.20721/) (2025)
- [Normalizing Flows with Iterative Denoising](https://mlsanity.com/paper/2604.20041/) (2026)
- [YoCausal: How Far is Video Generation from World Model? A Causality Perspective](https://mlsanity.com/paper/2605.30346/) (2026)
- [Chimera: Designing and Chinchilla-Scaling Hybrid Visual Diffusion Transformers](https://mlsanity.com/paper/2607.28611/) (2026)
- [Unified Neural Scaling Laws](https://mlsanity.com/paper/2605.26248/) (2026)
- [The FID Lottery: Quantifying Hidden Randomness in Generative-Model Evaluation](https://mlsanity.com/paper/2606.20536/) (2026)
- [Scaling Properties of Text Conditioning in Visual Generation](https://mlsanity.com/paper/2607.29679/) (2026)
- [Audio Generation Through Score-Based Generative Modeling: Design Principles and Implementation](https://mlsanity.com/paper/2506.08457/) (2025)
- [AMix-1: A Pathway to Test-Time Scalable Protein Foundation Model](https://mlsanity.com/paper/2507.08920/) (2025)

## Abstract (arXiv)

Diffusion transformers (DiT) have already achieved appealing synthesis and scaling properties in content recreation, e.g., image and video generation. However, scaling laws of DiT are less explored, which usually offer precise predictions regarding optimal model size and data requirements given a specific compute budget. Therefore, experiments across a broad range of compute budgets, from 1e17 to 6e18 FLOPs are conducted to confirm the existence of scaling laws in DiT for the first time. Concretely, the loss of pretraining DiT also follows a power-law relationship with the involved compute. Based on the scaling law, we can not only determine the optimal model size and required data but also accurately predict the text-to-image generation loss given a model with 1B parameters and a compute budget of 1e21 FLOPs. Additionally, we also demonstrate that the trend of pre-training loss matches the generation performances (e.g., FID), even across various datasets, which complements the mapping from compute to synthesis quality and thus provides a predictable benchmark that assesses model performance and data quality at a reduced cost.